Workflow and Tool Execution¶
The execute
command provides tools to execute CWL documents locally or on a remote server (soonishâ„¢).
Usage
Execution of CWL Files locally or on remote servers
Usage: s4n execute <COMMAND>
Commands:
local Runs CWL files locally using a custom runner or cwltool [aliases: l]
make-template Creates job file template for execution (e.g. inputs.yaml)
help Print this message or the help of the given subcommand(s)
Options:
-h, --help Print help
execute local
¶
There are two options for local execution. Using the CWL reference runner cwltool
which needs to be installed as extern dependency using pip install cwltool
or the SciWIn client runner which supports a large subset of CWL but lacks support containerization. You may ask yourself: why is there a custom runner? Because cwltool
only supports Windows using the Windows Subsystem for Linux (wsl) which is deactivated on many enterprise systems. The intention is to have a simple tool to test the generated CWL documents before sending them to the remote server.
The usage of the internal runner, which is the default one, is similar to the usage of cwltool
. It accepts the cwl file as first parameter and the inputs following at the end of the commands either as command line string or yaml file.
Usage
Runs CWL files locally using a custom runner or cwltool
Usage: s4n execute local [OPTIONS] <FILE> [ARGS]...
Arguments:
<FILE> CWL File to execute
[ARGS]... Other arguments provided to cwl file
Options:
-r, --runner <RUNNER> Choose your cwl runner implementation [default: custom] [possible values: cwltool, custom]
--outdir <OUT_DIR> A path to output resulting files to
--quiet Runner does not print to stdout
-h, --help Print help
execute remote
¶
Not yet implemented
execute make-template
¶
s4n execute make-template
is able to create a dummy CWL job file (e.g. inputs.yaml) that can be used as a template for an upoming execution of CWL.